Emanuela Lupo

PET and PVC Separation with Hyperspectral Imagery

Traditional plants for plastic separation in homogeneous products employ material physical properties (for instance density). Due to the small intervals of variability of different polymer properties, the output quality may not be adequate. Sensing technologies based on hyperspectral imaging have been introduced in order to classify materials and to increase the quality of recycled products, which have to comply with specific standards determined by industrial applications. This paper presents the results of the characterization of two different plastic polymers—polyethylene terephthalate (PET) and polyvinyl chloride (PVC)—in different phases of their life cycle (primary raw materials, urban and urban-assimilated waste and secondary raw materials) to show the contribution of hyperspectral sensors in the field of material recycling. This is accomplished via near-infrared (900–1700 nm) reflectance spectra extracted from hyperspectral images acquired with a two-linear-spectrometer apparatus. Results have shown that a rapid and reliable identification of PET and PVC can be achieved by using a simple two near-infrared wavelength operator coupled to an analysis of reflectance spectra. This resulted in 100% classification accuracy. A sensor based on this identification method appears suitable and inexpensive to build and provides the necessary speed and performance required by the recycling industry.

Investigation on an innovative technology for wet separation of plastic wastes.

This paper presents an original device for the separation of plastic polymers from mixtures. Due to the combination of a characteristic flow pattern developing within the apparatus and density, shape and size differences among two or more polymers, this device allows their separation into two products, one collected within the instrument and the other one expelled through its outlet ducts. Experimental tests have been conducted to investigate the effectiveness of the apparatus, using two geometric arrangements, nine hydraulic configurations and three selections of polymers at three stages of a material life cycle. Tests with samples composed of a single typology of polymer have been used to understand the interaction between the particles and the carrying fluid within the apparatus in different hydraulic configurations and geometric arrangements. Multi-material tests are essential to simulate the real conditions in an industrial recycling plant. The separation results have been evaluated in terms of grade and recovery of a useful material. Under the proper hydraulic configurations, the experimentation showed that it is possible to produce an almost pure concentrate of Polyethylene Terephthalate (PET) from a mixture of 85% PET and 15% Polycarbonate (PC) (concentrate grade and recovery equal to 99.5% and 95.1%) and a mixture of 85% PET and 15% Polyvinyl Chloride (PVC) (concentrate grade and recovery equal to 97.9% and 100.0%). It is further demonstrated that almost pure concentrates of PVC and PC can be produced from a mixture of 85% PVC and 15% PC (PVC grade and recovery equal to 99.9% and 99.7%) and a mixture of 85% PC and 15% PVC (PC grade and recovery equal to 99.0% and 99.5%).

Hyperspectral proximal sensing of Salix Alba trees in the Sacco river valley (Latium, Italy).

Recent developments in hardware and software have increased the possibilities and reduced the costs of hyperspectral proximal sensing. Through the analysis of high resolution spectroscopic measurements at the laboratory or field scales, this monitoring technique is suitable for quantitative estimates of biochemical and biophysical variables related to the physiological state of vegetation. Two systems for hyperspectral imaging have been designed and developed at DICEA-Sapienza University of Rome, one based on the use of spectrometers, the other on tunable interference filters. Both systems provide a high spectral and spatial resolution with low weight, power consumption and cost. This paper describes the set-up of the tunable filter platform and its application to the investigation of the environmental status of the region crossed by the Sacco river (Latium, Italy). This was achieved by analyzing the spectral response given by tree samples, with roots partly or wholly submerged in the river, located upstream and downstream of an industrial area affected by contamination. Data acquired is represented as reflectance indices as well as reflectance values. Broadband and narrowband indices based on pigment content and carotenoids vs. chlorophyll content suggest tree samples located upstream of the contaminated area are ‘healthier’ than those downstream.

Separation of plastic waste via the hydraulic separator Multidune under different geometric configurations

The recovery of high-quality plastic materials is becoming an increasingly challenging issue for the recycling sector. Technologies for plastic recycling have to guarantee high-quality secondary raw material, complying with specific standards, for use in industrial applications. The variability in waste plastics does not always correspond to evident differences in physical characteristics, making traditional methodologies ineffective for plastic separation. The Multidune separator is a hydraulic channel allowing the sorting of solid particles on the basis of differential transport mechanisms by generating particular fluid dynamic conditions due to its geometric configuration and operational settings. In this paper, the fluid dynamic conditions were investigated by an image analysis technique, allowing the reconstruction of velocity fields generated inside the Multidune, considering two different geometric configurations of the device, Configuration A and Configuration B. Furthermore, tests on mono- and bi-material samples were completed with varying operational conditions under both configurations. In both series of experiments, the bi-material samples were composed of differing proportions (85% vs. 15%) to simulate real conditions in an industrial plant for the purifying of a useful fraction from a contaminating fraction. The separation results were evaluated in terms of grade and recovery of the useful fraction.

Hydraulic separation of plastic wastes: Analysis of liquid-solid interaction

The separation of plastic wastes in mechanical recycling plants is the process that ensures high-quality secondary raw materials. An innovative device employing a wet technology for particle separation is presented in this work. Due to the combination of the characteristic flow pattern developing within the apparatus and density, shape and size differences among two or more polymers, it allows their separation into two products, one collected within the instrument and the other one expelled through its outlet ducts. The kinematic investigation of the fluid flowing within the apparatus seeded with a passive tracer was conducted via image analysis for different hydraulic configurations. The two-dimensional turbulent kinetic energy results strictly connected to the apparatus separation efficacy. Image analysis was also employed to study the behaviour of mixtures of passive tracer and plastic particles with different physical characteristics in order to understand the coupling regime between fluid and solid phases. The two-dimensional turbulent kinetic energy analysis turned out to be fundamental to this aim. For the tested operating conditions, two-way coupling takes place, i.e., the fluid exerts an influence on the plastic particle and the opposite occurs too. Image analysis confirms the outcomes from the investigation of the two-phase flow via non-dimensional numbers (particle Reynolds number, Stokes number and solid phase volume fraction).

Hyperspectral Monitoring of Green Roof Vegetation Health State in Sub-Mediterranean Climate: Preliminary Results

In urban and industrial environments, the constant increase of impermeable surfaces has produced drastic changes in the natural hydrological cycle. Decreasing green areas not only produce negative effects from a hydrological-hydraulic perspective, but also from an energy point of view, modifying the urban microclimate and generating, as shown in the literature, heat islands in our cities. In this context, green infrastructures may represent an environmental compensation action that can be used to re-equilibrate the hydrological and energy balance and reduce the impact of pollutant load on receiving water bodies. To ensure that a green infrastructure will work properly, vegetated areas have to be continuously monitored to verify their health state. This paper presents a ground spectroscopy monitoring survey of a green roof installed at the University of Calabria fulfilled via the acquisition and analysis of hyperspectral data. This study is part of a larger research project financed by European Structural funds aimed at understanding the influence of green roofs on rainwater management and energy consumption for air conditioning in the Mediterranean area. Reflectance values were acquired with a field-portable spectroradiometer that operates in the range of wavelengths 350–2500 nm. The survey was carried out during the time period November 2014–June 2015 and data were acquired weekly. Climatic, thermo-physical, hydrological and hydraulic quantities were acquired as well and related to spectral data. Broadband and narrowband spectral indices, related to chlorophyll content and to chlorophyll–carotenoid ratio, were computed. The two narrowband indices NDVI705 and SIPI turned out to be the most representative indices to detect the plant health status.